Scholarlink

Early detection is essential for improved treatment results since oral squamous cell carcinoma (OSCC) has a substantial influence on patient survival and quality of life. To distinguish OSCC from normal tissue in histological images, this study uses deep learning techniques. To enhance feature extraction and classification accuracy, we incorporated attention mechanisms into state-of-the-art models. The Attention-Based VGG16 model showed improved effectiveness in OSCC detection across the architectures, outperforming both VGG16 + ConvMixer and Attention-Based VGG16 + ConvMixer. Our results highlight the significance of attention modules in medical image processing and indicate that Multi-Head Attention may significantly enhance precision cancer diagnostic performance when integrated into VGG16 and carefully tuned.

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Optimized Histopathological Analysis Using Deep Learning for Early Detection of Oral Cancer

  • 会议论文

  • 发表日期：

    2026年

查看全文
  • G. Mahesh,
  • Shiva Shankar Reddy,
  • V. V. R. Maheswara Rao,
  • N. Silpa,
  • K. Amrutha,
  • K. V. S. S. R. Murthy

摘要

Early detection is essential for improved treatment results since oral squamous cell carcinoma (OSCC) has a substantial influence on patient survival and quality of life. To distinguish OSCC from normal tissue in histological images, this study uses deep learning techniques. To enhance feature extraction and classification accuracy, we incorporated attention mechanisms into state-of-the-art models. The Attention-Based VGG16 model showed improved effectiveness in OSCC detection across the architectures, outperforming both VGG16 + ConvMixer and Attention-Based VGG16 + ConvMixer. Our results highlight the significance of attention modules in medical image processing and indicate that Multi-Head Attention may significantly enhance precision cancer diagnostic performance when integrated into VGG16 and carefully tuned.